HIGHLIGHTS
- who: . et al. from the Manipal Institute of Technology, Karnataka, India have published the Article: Bayesian graph convolutional neural networks via tempered MCMC, in the Journal: (JOURNAL)
- what: The authors evaluate distinct features of Bayesian graph convolutional neural_networks (GCNNs) in terms of computational efficiency of parallel tempering MCMC sampling, effect of Langevin-gradient proposal distribution, and prediction accuracy for established benchmark datasets. The authors provide comparison with SGD in the following section. Prior literature does not always report the same summaries (best, mean, standard deviation) that the authors do, and so a direct comparision is . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.